First pass visualizations of VL cases in Brazil

Check distribution, look for weird outliers

The numbers in the upper right of each panel show the total number of cases

Over space

Total cases between 2007 and 2018 by municipality

Over space and time

Total cases by year (summed over months). Struggling to get this to print bigger

Total cases by month (summed over years). Struggling to get this to print bigger

Coarse correlations between predictors and cases

Total Precipitation, each data point is the number of VL cases in a given month, year, and municipality

Broken down by municipality to look for variation in the effects of the predictor over space

Human Population and Municipality Size, each data point is the number of VL cases in a given month, year, and municipality

Median NDVI, each data point is the number of VL cases in a given month, year, and municipality

Mean Air Temp, each data point is the number of VL cases in a given month, year, and municipality

Broken down by municipality to look for variation in the effects of the predictor over space

GDP over time and correlation to cases

Some notes:

In most municipalities seasonality is minimal
In most municipalities there is relatively little trend over years (some are decreasing, some are increasing?)
Is there a discernable pattern in cases over space over years (e.g., movement westward or…)?
Is there a discernable pattern in cases over months across space?
If using a random effect framework, how much goes into the random effects? (1 | munip) or much more (1 + year + beta | munip) ?